IntroductionStimulus-dependent protein scaffold formation is requiredfor many intracellular signaling pathways, including thoseassociated with protein kinases A and C (PKA and PKC),receptor tyrosine kinases, G-protein-coupled receptors, andintegrins (Carpenter, 2000; Feliciello et al., 2001; Miller andLefkowitz, 2001; Pryciak, 2001; Sim and Scott, 1999; Zamirand Geiger, 2001). Extracellular-signal-regulated kinases 1 and2 (ERK1/2), and other members of the mitogen-activatedprotein kinase (MAPK) family are downstream in many ofthese signaling cascades and are also controlled by intracellularcompartmentalization through interactions with scaffoldingproteins (Ge et al., 2003; Pouysségur and Lenormand, 2003).Among the many effector pathways regulated by labile scaffoldformation and localized signaling are the membrane andcytoskeleton rearrangements involved in cell translocation andcontractility (Carpenter, 2000; Feliciello et al., 2001; Ge et al.,2003).

Although the signaling pathways that control contractilityin differentiated smooth-muscle cells remain to be fullyelucidated, both myosin-linked and actin-linked mechanismsare involved (Harnett and Biancani, 2003; Gerthoffer andGunst, 2001; Pfitzer, 2001; Wier and Morgan, 2003; Gusev,2001; Morgan and Gangopadhyay, 2001; Somlyo and Somlyo,2003). Calcium ions bound to calmodulin regulate thecontractility of smooth-muscle myosin II through activationof myosin-light-chain kinase (MLCK). However, tonic

contractility in vascular smooth muscle does not necessarilycorrelate with myosin-light-chain phosphorylation (Jiang andMorgan, 1989; Moreland et al., 1992). Additional proposedregulatory targets include agonist-mediated activation of PKCand ERK1/2, and phosphorylation of the actin-binding proteincaldesmon.

Calponin (CaP) was originally isolated from chicken gizzard(Takahashi et al., 1986) and has been proposed as an adapterprotein in the actin-linked, agonist-induced regulation ofcontractility in smooth muscle (Gusev, 2001; Morgan andGangopadhyay, 2001; Small and Gimona, 1998). Caldesmonand CaP compete with each other for binding to actin filamentsin vitro and localize differently in vivo (for reviews, see Gusev,2001; Small and Gimona, 1998; Wang, 2001). Caldesmon isfound predominantly with actin filaments in the contractileapparatus, whereas CaP can localize with filamentous actin (F-actin) in the surrounding cortical cytoskeleton and membraneskeleton (Khalil et al., 1995; Parker et al., 1998). Thus, CaP isbetter situated to mediating membrane-associated contractileprocesses.

Antisense-mediated knockdown of CaP inhibits smooth-muscle contractility and ERK1/2 activation (Je et al., 2001).CaP binds directly to PKC and ERK1/2 (Leinweber et al.,1999; Leinweber et al., 2000). CaP also binds to F-actin andinhibits the actin-activated Mg ATPase of smooth-musclemyosin in vitro, an inhibition that can be reversed byphosphorylation of CaP (for a review, see Winder et al., 1998).

5043

The mechanisms by which protein kinase C (PKC) andextracellular-signal-regulated kinases (ERK1/2) governsmooth-muscle contractility remain unclear. Calponin(CaP), an actin-binding protein and PKC substrate,mediates signaling through ERK1/2. We report here thatCaP sequences containing the CaP homology (CH) domainbind to the C-terminal 251 amino acids of smooth-musclearchvillin (SmAV), a new splice variant of supervillin,which is a known actin- and myosin-II-binding protein. TheCaP-SmAV interaction is demonstrated by reciprocal yeasttwo-hybrid and blot-overlay assays and by colocalization inCOS-7 cells. In differentiated smooth muscle, endogenousSmAV and CaP co-fractionate and co-translocate to the cellcortex after stimulation by agonist. Antisense knockdown

of SmAV in tissue inhibits both the activation of ERK1/2and contractions stimulated by either agonist or PKCactivation. This ERK1/2 signaling and contractile defect issimilar to that observed in CaP knockdown experiments.In A7r5 smooth-muscle cells, PKC activation by phorbolesters induces the reorganization of endogenous,membrane-localized SmAV and microfilament-associatedCaP into podosome-like structures that also contain F-actin, nonmuscle myosin IIB and ERK1/2. These resultsindicate that SmAV contributes to the regulation ofcontractility through a CaP-mediated signaling pathway,involving PKC activation and phosphorylation of ERK1/2.

Key words: Supervillin, Calponin, PKC, ERK, Podosome

Summary

Smooth muscle archvillin: a novel regulator ofsignaling and contractility in vascular smooth muscleSamudra S. Gangopadhyay 1, Norio Takizawa 2, Cynthia Gallant 1, Amy L. Barber 1, Hyun-Dong Je 1,Tara C. Smith 2, Elizabeth J. Luna 2 and Kathleen G. Morgan 1,3,*1Boston Biomedical Research Institute, 64 Grove Street, Watertown, MA 02472, USA2Cell Dynamics Group, Department of Cell Biology, University of Massachusetts Medical School, 55 Lake Avenue, Worcester, MA 01605, USA3Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215, USA*Author for correspondence (e-mail: morgan@bbri.org)

Accepted 24 June 2004Journal of Cell Science 117, 5043-5057 Published by The Company of Biologists 2004doi:10.1242/jcs.01378

Research Article

5044

Upon agonist stimulation in vivo, CaP translocates andimmunoprecipitates with ERK1/2 and PKC in differentiatedsmooth muscle (Menice et al., 1997).

Archvillin is a ~250-kDa membrane skeleton protein thatlocalizes to cortical punctae in differentiating C2C12 cells and,with dystrophin, to costameres, which are sites of forcetransduction at the plasma membrane in striated muscle (Ohet al., 2003). Archvillin is a larger isoform of nonmusclesupervillin, which binds tightly to membranes, F-actin andsmooth-muscle and nonmuscle myosin II (Chen et al., 2003;Pestonjamasp et al., 1995; Pestonjamasp et al., 1997).Supervillin is isolated with signaling and lipid-raft-organizingproteins from neutrophil plasma membranes, and has beenimplicated in the regulation of cell adhesion (Nebl et al., 2002;Pestonjamasp et al., 1997). Supervillin and archvillin bothcontain sequences that activate the transcriptional activityof the androgen receptor (Ting et al., 2002), and nuclearlocalization has been observed (Oh et al., 2003; Pestonjamaspet al., 1997; Ting et al., 2002). Overexpression of the muscle-specific archvillin sequences in differentiating C2C12 cellsinhibits myotube formation, consistent with a dominant-negative effect during early myogenesis (Oh et al., 2003).

In this paper, we identify smooth-muscle archvillin (SmAV)as a unique isoform of supervillin (SV) that interacts with CaP.SmAV is most similar to, but is distinct from, skeletal-musclearchvillin. We discovered the CaP-SmAV association by yeasttwo-hybrid analyses with CaP as bait. The purposes of thepresent study were: (1) to confirm the CaP-SmAV interaction;and (2) to test whether SmAV contributes to the regulation ofcontractility through the CaP-mediated signaling pathway,which requires PKC activation and phosphorylation ofERK1/2.

The interaction was confirmed both by reciprocal blotoverlays and by colocalization of CaP and C-terminalsequences of supervillin (c-SV) that lack the actin- andmyosin-II-binding sites. CaP and c-SV colocalize to or near theplasma membrane after overexpression in COS-7 cells andbefore and after permeabilization with detergent. CaP andSmAV colocalize at endogenous levels to the cortices of freshlyisolated smooth-muscle cells, after agonist stimulation.Antisense knockdown of SmAV expression decreases PKC-activated contractility and phosphorylation of ERK1/2,signaling effects similar to those observed previously for a CaPknockdown in this system (Je et al., 2001). SmAV, CaP andERK1/2 localize with F-actin and nonmuscle myosin IIB toPKC-induced ‘podosome-like’ structures in A7r5 smooth-muscle cells, suggesting that these structures represent sites ofCaP- and SmAV-mediated signaling. Taken together, ourresults show that SmAV is an essential regulator of contractilityin differentiated smooth muscle and suggest that SmAV acts inconcert with CaP, PKC and ERK1/2 in the plasma membranecortex subsequent to agonist stimulation.

Materials and MethodsYeast two-hybrid screeningCaP cDNA (Je et al., 2001) was cloned into the bait vector pBD-GAL4 Cam (Stratagene, La Jolla, CA). A HybriZAP 2.1 two-hybridferret aorta cDNA library was constructed as described (Je et al.,2001), excised from phage and transformed into yeast strain YRG-2containing the bait vector. Transformants with target plasmids were

screened for histidine prototrophy and for β-galactosidase activityaccording to the manufacturer’s instructions (Stratagene). Interactingplasmid DNA (pAD-GAL4-2.1-c-SmAV) was isolated from apositively interacting colony and identified by DNA sequencing andBLAST search. To confirm the interaction of CaP and the C-terminusof SmAV, the bait and prey were exchanged. CaP cDNA was clonedinto pAD-GAL4-2.1 vector (pAD-GAL4-2.1-CaP) and the cDNA ofc-SmAV was cloned into pBD-GAL4 Cam vector (pBD-GAL4 Cam-c-SmAV). The deletion constructs of CaP were made by PCRamplification of the cDNA residues corresponding to amino acids 1-136 and 1-163. They were cloned into the pAD-GAL4-2.1 vector(pAD-GAL4-2.1-Cap136 and pAD-GAL4-2.1-CaP163, respectively).

Reverse-transcription PCRFor reverse-transcription PCR (RT-PCR) cDNA was made fromrandom-primed total ferret aorta RNA (1 µg) generated with theAdvantage RT-for-PCR kit (BD Biosciences Clontech, Palo Alto, CA)and diluted to 100 µl. PCR amplification was performed with two setsof primers: MSV-F1 (5′-YTGCACTYCCTAAAGYTGCAGAAT-TAAGAMAAAT-3 ′) with CRATY-R (5′-ACACGTAAACTTCAC-TACCAAAGTCAAACACCAG-3′); and C (5′-GTCCCTGATGAC-GACTACTGGGGG-3′) with D (5′-AACTGCTGAGGATGAA-CAGGCGGG-3′). Amplifications were performed using a PTC-100thermocycler (MJ Research, Waltham, MA) and a program as follows:92°C for 2 minutes (1 cycle); 92°C for 30 seconds, 68°C for 8 minutes(5 cycles); 92°C for 30 seconds, 63°C for 30 seconds, 70°C for 8minutes (5 cycles); 92°C for 30 seconds, 57°C for 30 seconds, 70°Cfor 8 minutes (30 cycles); and 70°C for 10 minutes (1 cycle). Thereaction was then held at 10°C overnight. The amplified products werecloned and sequenced in both directions.

Expression constructsA glutathione-S-transferase (GST) expression construct encoding theC-terminal 251 amino acids of SmAV (GST-c-SmAV) was generatedby directionally ligating the two-hybrid target vector into theEcoRI/XhoI sites of pGEX-6P-1 (Amersham Pharmacia Biotech,Arlington Heights, IL). Similarly, a chimera of GST and the C-terminal 510 amino acids of bovine supervillin (accession numberAF025996; GST-c-BSV) was generated by ligating a partial cDNAinto the SmaI/NotI sites of pGEX-6P-3. Recombinant His-taggedferret h1 CaP was constructed by PCR amplification of the openreading frame (Je et al., 2001) with oligonucleotide primerscontaining BamHI (sense) and HindIII (antisense) restriction sites.Amplified products were inserted in frame into BamHI/HindIII-digested pET 30b vector (Novagen, Milwaukee, WI) for bacterialexpression and into the pCDNA4/HisMax TOPO vector (InvitrogenLife Technologies, Baltimore, MD) for expression in COS-7 cells.Construction of EGFP-c-SV (EGFP-SV1010-1792) has beendescribed (Wulfkuhle et al., 1999).

Purified proteinsRecombinant His-tagged ferret h1 CaP was expressed after IPTGinduction of BL21 (DE3) Escherichia coliand purified using a Talonspin column (BD Biosciences) in the presence of 8 M urea. Purifiedprotein was dialysed against 20 mM Tris HCl, pH 8.0, 100 mM NaCl,10% glycerol and clarified by centrifugation. GST fusion proteinswere expressed similarly in BL21(DE3)LysS cells and affinitypurified on glutathione-Sepharose (Swaffield and Johnston, 2003).Proteins were eluted with 10 mM glutathione and dialysed against 20mM MOPS, pH 7.5, 2 µg ml–1 leupeptin, 60 mM KCl, 0.1% β-mercaptoethanol and 0.1 mM EGTA. Samples were frozen quickly inliquid nitrogen and stored at –80°C. Turkey-gizzard CaP purificationand subsequent digestion with chymotrypsin were performed aspreviously described (Leinweber et al., 2000).

Journal of Cell Science 117 (21)

5045Smooth muscle archvillin

AntibodiesMouse monoclonal anti-h1CaP antibody (Sigma Chemical, St Louis,MO) was used at 1:5000 for COS-7 cell experiments and at 1:20,000with aortic cells. The rabbit polyclonal antibody against the first 340residues of human supervillin (anti-H340) (Oh et al., 2003) was usedat 1:10,000 dilution for immunoblotting and cell staining. Other rabbitpolyclonal antibodies were used as follows: anti-green-fluorescent-protein (anti-GFP) (Santa Cruz Biotechnology, Santa Cruz, CA),1:2000; anti-p44/42 MAPK/ERK1&2 (Cell Signaling Technology,Beverly, MA), 1:500; and anti-phospho-p44/42 MAPK/pERK1&2(Cell Signaling Technology), 1:2000 dilution. Secondary antibodiesused in immunofluorescence microscopy were from Molecular Probes(Eugene, OR) as follows: Oregon-green-conjugated goat anti-mouseand Rhodamine-Red-X-conjugated goat anti-rabbit for smooth-muscle cells; Rhodamine-Red-X-conjugated goat anti-mouse andOregon-green-conjugated goat anti-rabbit for COS-7 cells; and Alexa-488-conjugated goat anti-rabbit and Texas-Red-conjugated goat anti-mouse for A7r5 cells.

COS-7 cellsCOS-7 cells were grown to 90-95% confluence in six-well plates withDulbecco’s modified Eagle’s medium containing 10% fetal bovineserum. Each well was transfected with 10 µg plasmid DNA (10 µgeach for double transfections) in Lipofectamine 2000 reagent(Invitrogen). After a 48-hour incubation at 37°C, cells weretrypsinized, plated onto coverslips and incubated for another 4 hoursat 37°C. Coverslips were washed with PBS and fixed with 2%paraformaldehyde, permeabilized with 0.1% Triton X-100, blockedwith 10% goat serum and reacted with primary and secondaryantibodies. In some experiments, cells were pre-permeabilized with0.1% or 0.5% Triton X-100 before fixation.

Ferret aorta cell isolationFerrets (Marshall Farms, North Rose, NY) were euthanized by anoverdose of chloroform and abdominal aortas were excised quickly toa dish containing oxygenated physiological saline solution (PSS; 120mM NaCl, 5.9 mM KCl, 25 mM NaHCO3, 11.5 mM dextrose, 2.5mM CaCl2, 1.2 mM MgCl2 and 1.2 mM NaH2PO4, pH 7.4). Allprocedures were approved by the Boston Biomedical ResearchInstitute Animal Care and Use Committee. Single aorta cells wereisolated using a variation of a previously described method (Meniceet al., 1997). Each 100 mg of aorta (wet weight) were digested withmedium A, which is composed of 1.2 mg CLS 2 collagenase (type II,390 U mg–1, lot number MIE4817; Worthington Biochemical,Lakewood, NJ) and 1.6 mg elastase (4.0 U mg–1; BoehringerMannheim, Indianapolis, IN) in a final total volume of 7.5 ml Ca- andMg-free Hanks’ balanced salt solution with 0.2% bovine serumalbumin (BSA). Tissue pieces were incubated in a shaking water bathat 34°C for 90 minutes, filtered on a nylon mesh, rinsed and incubatedfor 20 minutes in digestion medium B (medium A with only 0.74 mgelastase). Tissue pieces were filtered, rinsed and reincubated indigestion medium C [medium B plus 0.5 mg of 5000 U soybeantrypsin inhibitor (type II-S; Sigma)]. For all experiments, cells on onecoverslip were tested to confirm that they shortened in response tophenylephrine. Cells were fixed and stained for immunofluorescenceas described for COS-7 cells. To prevent cell shortening duringphenylephrine-induced stimulation, both stimulated and unstimulatedcells were placed in a Hanks’ solution to which 300 mM sucrose wasadded to increase tonicity, as previously described (Khalil et al., 1995;Parker et al., 1998). This procedure has been shown not to interferewith signal transduction in smooth-muscle cells, as measured byagonist-induced increases in intracellular calcium concentration andby translocation of CaP and PKC, and is thought to preventcontraction directly at the cross-bridge level (for review, see Parker etal., 1998).

ImagingCOS-7 and aorta cells

Images were obtained with a Kr/Ar laser (Radiance 2000, BioRadLaboratories) scanning confocal microscope equipped with Nikon603 (NA 1.4) or 403 (NA 1.4) oil immersion objective. Images wererecorded and quantified with Laser Sharp 2000 (BioRad Laboratories)for Windows NT. Ratio analyses of fluorescent intensity distributionswere performed as described (Khalil et al., 1994). The ‘surface’fluorescence value was defined as the peak pixel intensity over theouter cell surface, and the ‘cytoplasmic’ value was taken as the peakvalue in the core of the cell (see Fig. 8, inset). Each ratio was the meanof values from three line scans from non-nuclear areas of each cell.

A7r5 cellsA7r5 cells were stimulated for 2 hours with 1 µM phorbol dibutyrate(PDBu; Sigma), washed once with pre-warmed PBS, pH 7.4, andfixed in 4% paraformaldehyde. Cells were permeabilized for 10minutes with 0.1% Triton X-100 and blocked for 30 minutes in 1%BSA, 0.5% Tween-20, 0.02% sodium azide in PBS. Endogenouslevels of SmAV, myosin IIB, ERK1/2, and CaP were visualized with1:100 dilutions of rabbit polyclonal anti-H340 (Nebl et al., 2002),rabbit polyclonal anti-myosin IIB (Babco-Covance), rabbit polyclonalanti-ERK1/2 (UBI) and/or mouse monoclonal anti-CaP (clone hCaP;Sigma), respectively. F-actin was stained with a 1:1000 dilution ofTexas-Red-conjugated phalloidin (Molecular Probes). Images wereobtained with OpenLab software (Improvision, Lexington,MA) on aZeiss Axioskop fluorescence microscope.

Colocalization analysisThe degree of colocalization between pairs of stained proteins inCOS-7 cells and aorta smooth-muscle cells was quantified by amodification of a previously published method (Parker et al., 1998)and by using LaserPix image software (BioRad Laboratories). Imageswere thresholded to eliminate background signal outside cellboundaries. A colocalization coefficient between red and greenfluorophores (CR–G) was computed as,

where the denominator is the sum of red pixels intensities greater thanthe threshold and the numerator is the sum of the intensities of redpixels that are greater than threshold and that also have a greencomponent. The amount of reverse colocalization [i.e. (CG–R)] wascomputed in a similar manner by replacing the denominator of Eqn 1with the sum of intensities of green pixels that are greater than thethreshold. The colocalization coefficients are, by definition, between0 and 1. A value of 0 indicates no colocalization and a value of 1indicates complete colocalization.

The pixel intensities in each image were randomized by reading thepixel information within each image into a mathematical matrix. Theelements of each matrix were then assigned to new locations withinthe matrix using a random number generator. After converting therandomized matrices back into images, the colocalization coefficientsbetween the pair of random images were computed. A comparison ofthe colocalization coefficients for the random images to that for theoriginal images allows an assessment of the degree to whichcolocalization between a pair of fluorophores was due to chance.

Tissue fractionationAortic ferret tissue was equilibrated for 1 hour at room temperaturein oxygenated (95% O2, 5% CO2) PSS. Viability was tested with adepolarizing solution consisting of PSS in which 51 mM NaCl wasstoichiometrically replaced by KCl. The tissue was further incubated

C =R

iR(1)Σ

ΣR−Gi,coloc ,

5046

for 1 hour in PSS and quickly frozen in a dry-ice/acetone/dithiothreitol slurry in either the presence or the absence of 10 µMphenylephrine for 10 minutes. Tissue was pulverized andhomogenized at 4°C with 3 ml buffer I (20 mM Tris HCl, pH 7.5, 50mM NaCl, 250 mM sucrose, 10 mM dithiothreitol (DTT), 3 mMEGTA, 5 mM MgCl2, 1 mM ATP and protease inhibitors) (Je et al.,2001). Homogenized tissue was centrifuged at 165,000 g for 1 hourand the supernatant collected as the cytosolic fraction. The pellet wasshaken for 1 hour at 4°C in 3 ml buffer II (20 mM Tris HCl, pH 7.5,250 mM sucrose, 0.5% Triton X-100, 10 mM DTT, 3 mM EGTA, 5mM MgCl2, 1 mM ATP and protease inhibitors) and centrifuged at165,000 g for 1 hour. This supernatant was collected as the Triton-extractable (‘membrane’) fraction. The pellet was resuspended in 3ml buffer III [20 mM Tris HCl, pH 7.5, 250 mM sucrose, 0.5% TritonX-100, 1.2% sodium dodecyl sulfate (SDS), 10 mM DTT, 3 mMEGTA, 5 mM MgCl2, 1 mM ATP and protease inhibitors], extractedat 4°C for 1 hour and centrifuged at 165,000 g for 1 hour. This finalsupernatant was collected as the Triton-resistant (‘cytoskeletal’)fraction.

Western blotsTissue homogenate (25 µg) was loaded onto 4-15% Tris-HCl gradientgels (Bio-Rad Laboratories) and electroblotted onto PVDFmembranes (Millipore, Billerica, MA). Blots of purified proteins wereresolved on 4-15% gradient polyacrylamide gels (Laemmli, 1970) andelectrotransferred to PVDF (Schleicher & Shuell, Keene, NH). Blotswere blocked with 5% nonfat dried milk and incubated with primaryand secondary antibodies. Signals were visualized by enhancedchemiluminescence. PBDu-treated A7r5 cells (1×106) were treatedwith 10% trichloroacetic acid and centrifuged at 10,000 g for 10minutes at 4°C. The pellets were dissolved with 0.5 M NaHCO3 andTris base to adjust pH. Samples (1.5×105 cells per 4-mm lane) wereresolved on 6-20% gradient polyacrylamide gels and transferred tonitrocellulose, as above. Blots were washed with TBS containing

0.05% Tween 20 and incubated for 1 hour with primary antibodies atroom temperature. Signals were visualized with SuperSignal

WestPico (Pierce, Rockford, IL) after a 1-hour incubation withsecondary antibodies conjugated with horseradish peroxidase.

Overlay assayEqual amounts of proteins were resolved on SDS-polyacrylamidegels and transferred to PVDF membranes. Relative protein contentwas confirmed by naphthol blue black (NBB) staining of themembrane before incubation with the overlay solution. Blots wereblocked with 5% milk in TBS and incubated for 1 hour with 1.25µg ml–1 recombinant ferret CaP, GST-c-SmAV or GST in 154 mMNaCl, 10 mM Tris HCl, pH 7.4, 10 mg ml–1 BSA and proteaseinhibitors (Leinweber et al., 1999). Blots were fixed with 0.5%paraformaldehyde in PBS, neutralized with 2% glycine in PBS,washed six times with PBS and immunostained as for westernblots.

Antisense knockdown of SmAV expression in smooth-muscletissueExperiments were performed as previously described (Kim et al.,2000). The antisense oligonucleotide used was 5′-CCTTGC-CAGAGGTACACCTCAT-3′, corresponding to amino acidsHEVYLWQG (residues 1912-1919) of SmAV. A randomizedversion of this sequence (5′-CACTAGAGTCCAGCTACGCTCT-3′)served as the negative control. BLAST searches confirmed a lack ofhomology to any other relevant coding sequences. Nucleotides werephosphorothioated to minimize degradation and FITC tagged at their5′ ends. Oligonucleotides were loaded once each day until day 4, asdescribed previously (Je et al., 2001). Fluorescence microscopy atthe end of each experiment confirmed internalization of theoligonucleotide into all cells of the tissue (Je et al., 2001).Contractility of the muscle strips was measured daily in response to

Journal of Cell Science 117 (21)

Fig. 1. Identification of SmAV as a CaP interacting protein. (A) Interactions between SmAV and CaP in reciprocal yeast two-hybridassays. The growth of yeast host (YRG2) with plasmids or constructs in Trp-, Leu- and His-depleted selective medium. (1) Yeast Host-YRG2. (2) pBD-GAL4 Cam. (3) pBD-GAL4 Cam-CaP. (4) pBD-GAL4 Cam-CaP and pAD-GAL4-2.1-c-SmAV. (5) pBD-GAL4 Cam-c-SmAV. (6) pBD-GAL4 Cam-c-SmAV and pAD-GAL4-2.1-CaP. (7) pBD-GAL4 Cam-c-SmAV and pAD-GAL4-2.1-Cap136. (8) pBD-GAL4 Cam-c-SmAV and pAD-GAL4-2.1-CaP163. (B) CaP domains indicating the amino acid residue numbers (ferret sequence) and thedomains (Morgan and Gangopadhyay, 2001). Double-headed arrows indicate the deletion constructs used. (C) Expression of LacZ asdetected by blue color development with X-gal. Colonies of yeast cells (either host only or transfected with plasmid/constructs) arenumbered as in (A).

5047Smooth muscle archvillin

51 mM KCl. No significant decreases in the amplitude of contractionto KCl were observed over the 4 day course of the experiment. TheSmAV expression level and ERK1/2 phosphorylation levels weremeasured by densitometry of western blots from day 4 tissueextracts.

StatisticsUnless otherwise noted, significance of difference between means wastaken at P>0.05 by a two-tailed Student’s t test.

ResultsIdentification of SmAV as a CaP-interacting proteinWe used reciprocal yeast two-hybrid screens to identify a newinteraction partner for CaP. We initially used the cDNAencoding ferret (h1) basic CaP as bait in an untargetedscreen of a ferret aorta cDNA library. Out of thousands oftransformants, one specific interacting clone was identified andverified (Fig. 1A,C4). This plasmid contained an 862-bp insertencoding 251 amino acids. A BLAST search of protein

-CN-

A Calponin binding domain 1823-2073

First musclespecific insert 264-567

60% identity with human archvillin

88% identity with human archvillin

490-493

608-611

1049-1065

1212-1216

1 2 3

B

1 250 500 750 1000 1250 1500 1750 2073

GEL GEL GEL GEL GEL

VILLIN HEADPEICE

GEL GELSOLIN HOMOLOGY

GELSOLIN REPEAT

Smooth Muscle ArchVillin

SH2Y771

1312-1391

1447-1486

1515-1536

1674-1716

1907-1920

ERK binding siteP219

S248 Pro-depST kinasepST14-3-3-bind

T630

2039-2073

266-565

MUSCLE SPECIFIC INSERT

NLS

1823-2073 CALPONIN-BINDING DOMAIN

P774ERK binding

S913PKA

T1167PKCz

T1681CaMKII

490 608 1049 1212

F-ACTINBINDING

F-ACTIN BINDING

1-175

MYOSIN II BINDING

MYOSINBINDING

GEL

264-567

565-627629-707

984-1114

Fig. 2.Sequence comparisons ofSmAV with human supervillin andarchvillin. (A, top) The SmAVsequence. The muscle-specificsequence (insert 1) is in yellow (Ohet al., 2003) and the nuclear targetingsequences is in blue. The CaP-binding domain is represented by thered box. (A, bottom) The alignmentof N-terminal amino acid sequencesof ferret SmAV with those of humansupervillin and human archvillin.Identical residues are in black; thesecond muscle-specific insert inskeletal-muscle archvillin is in green.The three sequences in skeletal-muscle but not smooth-musclearchvillin are boxed and numbered.Residue numbers are given on theright. The SmAV sequence data areavailable from GenBank (accessionnumber AY380816). (B) Domainanalysis of SmAV, based onhomology with domains insupervillin (Chen et al., 2003) andmotif searching programs(http://scansite.mit.edu/,http://www.ncbi.nlm.nih.gov/).Domains are marked with residuenumbers. Sequences correspondingto the predicted binding sites formyosin II (lavender) and F-actin(brown) are shown. The CaP-bindingdomain is represented by the red box.

5048

databases (Altschul et al., 1997) showed 94% identity withhuman supervillin. In the reciprocal experiment, the supervillinhomolog in the bait vector also interacted with CaP as prey (Fig.1A,C6). A deletion mutant (CaP 1-163) lacking the CLIK motif(Gimona et al., 2003) of CaP (Fig. 1B) interacted with thesupervillin homolog (Fig. 1A,C8), as did a further deletionmutant (CaP 1-136) lacking both the TnI-Like domain and theCLIK motif but containing the single type 3 calponin homology(CH) domain (Gimona et al., 2002) of calponin (Fig. 1A,C7).The CH domain of CaP is of interest because it has also beenidentified as an ERK-binding domain (Leinweber et al., 1999).

The complete cDNA sequence for the supervillin-relatedprotein in ferret aorta revealed a new splice variant thatmight be characteristic of smooth muscle. The encodedprotein (Fig. 2) is predicted to exhibit a relative molecularmass of 231,457.96 and an isoelectric point of 6.44(http://us.expasy.org/tools/pi_tool.html). The CaP-bindingsequence initially identified by the yeast two-hybrid assay isthe C-terminal 251 residues (1823-2073). Alignment withpreviously described supervillin family members revealed 82%similarity (78% identity) with human archvillin and 78%similarity (74% identity) with human supervillin. The C-terminal 1416 residues are 88% identical to the correspondingsequence of both supervillin and archvillin.

Archvillin, a skeletal-muscle-specific isoform of supervillin,

contains two muscle-specific inserts encoded by four differentlyspliced exons in both human and murine tissues (Oh et al.,2003). The first muscle-specific archvillin insert is encoded byexons 3, 4 and 5 in skeletal muscle archvillin (Fig. 2A, yellow).The large exon 4 is conserved in the smooth-muscle archvillinprotein sequence (residues 264-567), except for a shorthypervariable region (Fig. 2A, region II). However, muscleinsert sequences encoded by exons 3 and 5 are absent (Fig. 2A,regions I,III), as is the second muscle-specific archvillin insert(Fig. 2A, green), which is encoded by exon 9 in murine andhuman archvillins (Oh et al., 2003). Because of the greatersimilarity to skeletal-muscle archvillin, we have named theprotein found in aortic smooth muscle SmAV for smooth-muscle archvillin. Thus, SmAV is intermediate in amino acidsequence between skeletal-muscle archvillin and non-musclesupervillin, presumably owing to alternative exon splicing.

The binding sites for myosin II and F-actin that have beenrecently mapped within the N-terminus of bovine supervillin(Chen et al., 2003) appear to be present and highly conservedwithin SmAV. SmAV contains a site from residues 1 to 175(Fig. 2B, pink) with 80% identity to the binding site for non-muscle and smooth-muscle myosin II in supervillin. Similarly,residues 565-627, 629-707 and 984-1114 (Fig. 2B, brown)contain residues with 97%, 93% and 91% identity, respectively,to the three actin-binding domains of supervillin.

Journal of Cell Science 117 (21)

Fig. 3.Overlay analysis of binding interactionsbetween CaP and supervillin/archvillin fragments.(A, left) Naphthol blue black (NBB) stain of totalprotein on blot before incubation with overlay solution.(A, right) Immunostain with anti-CaP antibody afterprocessing with overlay buffer containing 1.25 µg ml–1

CaP. Purified proteins and their corresponding overlaysignals in lanes are indicated with arrowheads. Allthree proteins were loaded at 0.2 µg per lane.(B) Quantitative analysis of results obtained in (A). Thechart shows the ratio of densitometry obtained from theoverlay assay to that from the NBB-stained blot. Theresults are from n=7 (GST-c-BSV), n=6 (GST-c-SmAV) and n=3 (GST) samples in each case. (C, left)NBB stain of total protein on blot before incubationwith overlay solution. (C, right) Immunostain withanti-GST antibody after processing with overlay buffercontaining 1.25 µg ml–1 GST-c-SmAV or GST only asindicated. Purified proteins and their correspondingoverlay signals in lanes are indicated with arrowheads.Both recombinant CaP and recombinant calmodulinwere at 3 µg per lane. (D) Quantitative analysis ofresults obtained in (C). The chart shows the ratio ofdensitometry obtained from the overlay assay to thatfrom the NBB-stained blot. The results are from n=5(CaP) and n=3 (CaM) samples for the GST-c-SmAVoverlay, and n=3 (CaP) and n=3 (CaM) samples for theGST overlay. (E) Overlay assay of GST-c-SmAV withchymotrypsin-digested fragments of CaP. (left) NBBstain of blot before incubation with overlay solution.(right) Immunostain with anti-GST antibody afterprocessing with overlay buffer containing 1.25 µg ml–1

GST-c-SmAV. Undigested (full-length) and fragmentsof CaP are indicated with arrowheads. (F) Quantitativeanalysis of results obtained in (E). The chart shows theratio of densitometry obtained from the overlay assayto that from the NBB-stained blot. The results are fromn=5 samples.

5049Smooth muscle archvillin

Analysis of the SmAV sequence by motif searching programsat high stringency (http://scansite.mit.edu/ and CD Search,NCBI, http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi)revealed several potentially significant regulatory sites. One ofthe two nuclear targeting sequences (Fig. 2A, blue) within thefirst insert in skeletal-muscle archvillin (Oh et al., 2003) isconserved in SmAV. C-Terminal sequences homologous to sixdomains in gelsolin and villin, characteristic of supervillins andarchvillins (Pestonjamasp et al., 1997), are also present inSmAV (Fig. 2B, red ovals and short rectangles). Additionally,several kinase-binding sites and phosphorylation sites arepredicted (Fig. 2B). Because CaP has been linked with PKC-and ERK-dependent signaling pathways, it is noteworthy thatthe SmAV sequence predicts the presence of two ERK-bindingsites and a motif consistent with phosphorylation by PKC anda proline-directed kinase.

Direct SmAV-CaP binding in vitroBinding between CaP and SmAV was confirmed by blotoverlay assays (Fig. 3). As is shown in Fig. 3A, recombinantCaP from the overlay buffer binds to blotted GST fusion

proteins containing the C-terminus of either bovine supervillin(GST-c-BSV, amino acids 1282-1792) or SmAV (GST-c-SmAV, amino acids 1822-2073). Neither protein bindsappreciably to equal amounts of GST (Fig. 3A). Quantitativedensitometry analysis (Fig. 3B) of the ratio of the CaPimmunostained band to that of the total protein band obtainedby NBB staining confirms that there is significant binding toboth GST-c-BSV and GST-c-SmAV but little binding to GSTalone. Specific binding was also observed in the reverseexperiment [i.e. overlays with GST-c-SmAV or GST in theoverlay buffer and CaP and calmodulin (for comparison) on theblot] (Fig. 3C). Even though as much or more calmodulin waspresent on the blot, more GST-c-SmAV (as visualized withanti-GST antibody) bound to CaP than to calmodulin (Fig. 3C).By contrast, with an equal amount of GST alone in the overlaybuffer, no binding was detected to either CaP or calmodulin onblots, even at long exposures (Fig. 3C). Quantitative analysis(Fig. 3D) of the ratio of the GST immunostained band to thatof the total protein band obtained by NBB staining confirmedthese observations.

To confirm the domain of CaP responsible for binding toSmAV, purified turkey-gizzard CaP was chymotrypsin digested

and fragments were identified by N-terminal sequencingand mass spectrometry (Leinweber et al., 2000). Thesefragments were used in overlay assays with GST-c-SmAVin the overlay buffer. The results indicate that GST-c-SmAV specifically binds to the 20 kDa fragment and tothe undigested full-length CaP, but that the overlay signalwith the 14 kDa fragment is almost undetectable (Fig. 3E).The quantitative analysis of the anti-GST-immunostainedbands normalized to those for total protein confirms theseresults (Fig. 3F). The 20 kDa fragment to which GST-c-SmAV bound contains residues 7-152 (Fig. 1B), whichencompass the single type-3 CaP-homology (CH) domainof CaP (Gimona et al., 2002).

Colocalization in COS-7 cellsTo verify the interaction between CaP andsupervillin family proteins in vivo, we tested forprotein colocalization in COS-7 cells, which aretransformed epithelial cells that are easilytransfectable with multiple plasmids. Wetransfected these cells with plasmids encodingHis-tagged ferret h1CaP and a previouslydescribed C-terminal fragment of bovinesupervillin tagged with enhanced greenfluorescent protein (EGFP-c-SV, amino acids1010-1792) (Wulfkuhle et al., 1999). This C-

Fig. 4.CaP and the supervillin C-terminus colocalizein COS-7 cells. Confocal images of immunostainedtransfected COS-7 cells. (A, left) Single expressionof the EGFP-tagged C-terminal 783 residues ofbovine supervillin (EGFP-c-SV). (A, right) Singleexpression of h1-CaP. (B) CaP (left) and EGFP-c-SV(middle) co-expressed in COS-7 cells. Merged imageis shown to the right; regions of signal overlapappear in yellow. (C) Co-expression of CaP andEGFP control. CaP was visualized by staining withan anti-CaP antibody and EGFP-c-SV was visualizedwith an anti-GFP antibody. Scale bar, 10 µm.

5050

terminal bovine supervillin fragment is 94% identical (96%homologous) to ferret SmAV in the region containing theputative CaP-binding site. We used this construct, which lacksthe N-terminal sequences of supervillin that interact with F-actin and myosin II (Chen et al., 2003; Wulfkuhle et al., 1999),because CaP also binds actin (Gimona et al., 2002). Thus,colocalization of EGFP-c-SV with CaP associated with actinfilaments can be informative in a way that colocalization withfull-length SmAV would not be.

CaP and EGFP-c-SV did colocalize in COS-7 cells (Fig. 4).In cells expressing EGFP-c-SV only, this protein wasconcentrated primarily in cytoplasmic punctae and at the cellperiphery (Fig. 4A, left, green). All nine randomly acquiredimages displayed this qualitative pattern. By contrast, cellsexpressing exogenous CaP by itself displayed both peripheral(12 out of 13 cells) and linear (13 out of 13 cells) staining (Fig.4A, right, red), consistent with CaP’s association with actin-containing microfilaments (Parker et al., 1998). In cellstransfected with both constructs, overlap in staining wasobserved at punctae reminiscent of cell-substrate adhesionplaques (12 out of 12 cells) (Fig. 4B, inset, arrows) and alongshort filaments in the cell interior (11 out of 12 cells) (Fig. 4B,inset, arrowheads) as well as at cell peripheries (12 out of 12cells). Calculation of colocalization coefficients confirms thatthere is a statistically very significant overlap of CaP withSmAV and of SmAV with CaP, compared with that for the sameimages when randomized (Table 1).

As a negative control, we co-transfected CaP and the emptyEGFP vector (Fig. 4C). As previously reported (Wulfkuhle etal., 1999), EGFP alone labeled the nucleus and littlecolocalization with CaP was observed. CaP again showed aprominent filamentous staining but EGFP does not (Fig. 4C).

To test more rigorously for colocalization of CaP and EGFP-c-SV, we pre-extracted doubly transfected COS-7 cells withTriton X-100 before fixation and staining. This procedurereleases proteins that are unbound or only loosely bound tocytoskeletal structures. As previously reported (Wulfkuhle etal., 1999), essentially all EGFP-c-SV expressed by itself inCOS-7 cells was released by pre-extraction with Triton X-100before cell fixation (Fig. 5A, right, green). By contrast,exogenously expressed CaP continued to exhibit a linearappearance in Triton-extracted cells (Fig. 5B, red). In cells

expressing both CaP and EGFP-c-SV, the extractability ofEGFP-c-SV with 0.5% Triton X-100 (Fig. 5C, right, green) andwith 0.1% Triton (not shown) was greatly decreased. In thesepre-extracted cells, CaP and EGFP-c-SV continued tocolocalize extensively in cytoplasmic punctae and along linearstructures (presumably microfilaments) even after Triton pre-extraction (Fig. 5C, right, overlap in yellow). These resultssuggest that CaP stabilizes the association of the supervillin C-terminus with the cytoskeleton and that EGFP-c-SV mightpotentiate the localization of CaP to cytoplasmic punctae.

Co-fractionation of CaP and SmAV from smooth muscleThe colocalization of overexpressed CaP and EGFP-c-SV inCOS-7 cells suggests an interaction between these two proteinsbut does not necessarily indicate that the endogenous proteinsinteract in differentiated smooth muscle. As one test of co-association of endogenous SmAV and CaP in differentiatedsmooth muscle tissue, we fractionated at high speed (165,000g) both unstimulated smooth-muscle tissue and tissuestimulated for 10 minutes with the agonist, phenylephrine (Fig.6). Three fractions were obtained: a cytosolic fraction (proteinsreleased by homogenization in the absence of detergent); aTriton-extractable (‘membrane’) fraction (proteins releasedwith 0.5% Triton X-100); and a Triton-insoluble(‘cytoskeletal’) fraction (proteins released with 1.2% SDS and0.5% Triton X-100). In both unstimulated and stimulatedtissues, SmAV and CaP co-fractionated in the cytoskeletalfraction (Fig. 6), consistent with an affinity for cytoskeletalstructures and/or Triton-insoluble membrane domains.

Agonist-dependent colocalizationTo define more precisely the localizations of endogenousSmAV and CaP in differentiated smooth-muscle cells, weperformed quantitative confocal imaging studies on freshlyisolated cells from the aorta of the ferret (Fig. 7). Using anantibody raised against the N-terminal 340 amino acids ofhuman supervillin (anti-H340) (Nebl et al., 2002), we detecteda single band at ~231 kDa in western blots of whole-cellhomogenates of ferret aorta (Fig. 7A). The size of this proteinis consistent with the size of SmAV predicted from the deduced

Journal of Cell Science 117 (21)

Table 1. Calponin and SmAV colocalize significantly in COS-7 cells, and agonist stimulation increases the colocalizationcoefficient in aorta cells

Cell type Protein Mean colocalization coefficient* s.e. P value

COS-7 cells Calponin-SmAV 0.79 0.02 –Random calponin-SmAV 0.41 0.02 0.0001SmAV-calponin 0.72 0.03 –Random SmAV-calponin 0.51 0.01 0.0003

Unstimulated FASMCs‡ SmAV-calponin 0.34 0.05 –Random SmAV-calponin 0.25 0.01 0.12Calponin-SmAV 0.58 0.05 –Random calponin-SmAV 0.37 0.03 0.02

FASMCs + phenylephrine SmAV-Calponin 0.51 0.20 –Random SmAV-calponin 0.17 0.01 0.0001Calponin-SmAV 0.80 0.02 –Random calponin-SmAV 0.23 0.02 0.0001

In each case, 10-25 cells were analysed.*The proportion of first protein that colocalizes with second protein.‡FASMCs, ferret-aorta smooth-muscle cells.

5051Smooth muscle archvillin

amino acid sequence and suggests that SmAV is the dominantmember of the supervillin/archvillin family in aorta smoothmuscle. The specificity of this antibody further suggests that itcan be used to image SmAV in this tissue.

In images of unstimulated smooth-muscle cells that had beenfreshly isolated from aortic tissue, we found, as has beenpreviously reported (Parker et al., 1998), that CaP is excludedfrom the nucleus, localizing primarily to linear structures (Fig.7B, left, green), presumably microfilament bundles (Parkeret al., 1998). These previous studies determined that CaPcolocalizes with actin in these bundles (Parker et al., 1998).SmAV also is excluded from the nucleus but, in contrast to CaP,is concentrated in a non-filamentous, more globular, patternthroughout the cell interior (Fig. 7B, middle, red). Somepunctate staining at the cell periphery is also detected(arrowheads). A merged image (Fig. 7B, right) of the CaP andSmAV signals shows little convincing overlap. Quantificationof colocalization coefficients confirmed that the degree ofcolocalization is lower than that in the COS-7 cells (Table 1).Furthermore, the proportion of SmAV that overlaps with CaPwas not statistically different from that due to chance, asmeasured in randomized versions of the same images. The

reverse comparison (i.e. the proportion of CaP that overlapswith SmAV), although low, was significantly greater than thatfor the randomized images. The difference from the randomimages is presumably due to the fact that the CaP is organizedinto discrete bundles overlying the more diffuse pattern ofSmAV.

Interestingly, a very different pattern emerges when the cellsare stimulated with the α-adrenergic agonist phenylephrine(Fig. 7C). As previously reported (Menice et al., 1997), theaddition of phenylephrine to ferret aortic cells causes CaP totranslocate to the cell periphery (Fig. 7C, left, green). Underthese conditions, the SmAV distribution also shifts towards thecell periphery (Fig. 7C, middle, red), where it distributes withCaP (Fig. 7C, right, yellow). This results in an increase in thecolocalization coefficients (Table 1). Furthermore, in thepresence of phenylephrine, both the overlap of SmAV on CaPand the overlap of CaP on SmAV are significantly greater thanthat seen in the corresponding randomized images. Ratioanalysis also confirms that the individual translocations of eachprotein to the cell periphery are significant (Fig. 8).

Fig. 5. Co-transfection with CaP preventsextraction of EGFP-c-SV by Triton X-100 fromCOS-7 cells. (A, left) Unextracted COS-7 cellsingly transfected with EGFP-c-SV.(A, right) COS-7 cell singly transfected withEGFP-c-SV and extracted with 0.5% Triton X-100 before fixation. (B, left) Unextracted COS-7cell singly transfected with CaP. (B, right) COS-7 cell singly transfected with CaP and extractedwith 0.5% Triton X-100 before fixation.(C, left) Unextracted COS-7 cell co-transfectedwith EGFP-c-SV (top) and CaP (middle).(C, right) COS-7 cell co-transfected with EGFP-c-SV (top) and CaP (middle) and extracted with 0.5% Triton X-100before fixation. Merged images shown at the bottom, with regions ofoverlap in yellow. Scale bar, 10 µm.

Fig. 6. Tissue fractionation of CaP and SmAV. Immunodetection ofSmAV (top) and CaP (bottom) in cytosolic, Triton-X-100-extractable(‘membrane’) and Triton-X-100-resistant (‘cytoskeletal’) fractionsfrom unstimulated tissue or from tissue stimulated for 10 minuteswith 10–5 M phenylephrine.

5052

CaP and SmAV co-translocate with similar time coursesTo confirm the statistical significance of the SmAVredistribution, we performed digital image analysis with analgorithm in which the surface fluorescence is divided by thefluorescence in the cell core (Khalil et al., 1994) (Fig. 8, inset).The phenylephrine-induced translocations of both SmAV andCaP were highly statistically significant (Fig. 8A,B). However,although the average ratios for SmAV and CaP were 1.35 and1.42, respectively, in unstimulated cells, stimulation caused theratio for SmAV to increase to 2.22, whereas the ratio for CaPincreased to 3.83. These results suggest that a larger proportionof the SmAV remains in the center of the cell.

To test whether SmAV and CaP co-translocate with a similartime course, ratios of SmAV distributions were determined asa function of time after phenylephrine addition (Fig. 8C).SmAV targeting to the cell periphery reaches a maximumwithin 4 minutes of stimulation, and persists through the 10minutes of observation. This time course is indistinguishablefrom that of CaP, as previously determined under similarexperimental conditions in the same cell type (Menice et al.,1997).

SmAV knockdown decreases PKC-dependentcontractility in vascular tissueTo investigate whether SmAV plays a regulatory role in the

contractile function of differentiated smooth muscle similarto that documented previously for CaP (Je et al., 2001), weused an antisense knockdown approach that preservescontractility in organ culture (Fig. 9). We achieved ~50%knockdown of SmAV expression in ferret aorta smooth-muscle tissue (Fig. 9A). By contrast, neither actin norERK1/2 total protein values changed with antisense treatment(data not shown). We have previously shown that this methodtransfects antisense oligonucleotides equally into all cellsacross the thickness of the tissue preparation (Je et al., 2001).Although no significant change was observed in themagnitude of contractility mediated by depolarization (Fig.9B), SmAV knockdown led to a significant decrease in themagnitude of contractions triggered by activation of PKCwith a phorbol ester (Fig. 9C). Treatment with a randomizedversion of the nucleotides used in the antisense sequence hadno significant effect on the magnitude of either type ofcontraction, as compared with sham-treated muscles (Fig.9B-D). Consistent with our previous observations that theproportion of phenylephrine-stimulated contraction thatpersists in a Ca-free buffer (about one-third of the totalcontraction) is specifically PKC dependent (Khalil et al.,1994; Menice et al., 1997), we found that calcium-independent, phenylephrine-induced contractions were alsosignificantly reduced by SmAV knockdown (Fig. 9D). Thisphenotype, with respect to both phorbol-ester- and

Journal of Cell Science 117 (21)

Fig. 7. CaP and SmAV colocalize in an agonist-dependent manner in differentiatedaortic smooth-muscle cells. (A) Western blot of aorta tissue homogenate probedwith anti-H340 antibody. Molecular weight standards are indicated on the left.(B) Confocal images of a freshly isolated ferret aorta cell fixed under controlconditions and labeled for endogenous levels of CaP (left, green) and SmAV(middle, red). Merged images are shown on the right. (C) Confocal images of a freshly isolated ferret aorta cell fixed in the presence of 10–5 Mphenylephrine for 10 minutes and labeled for CaP and SmAV. Merged images are on the right, with signal overlap in yellow. Scale bar, 10 µm.

5053Smooth muscle archvillin

Ca-independent phenylephrine-stimulated contractions, isessentially identical to that observed for CaP knockdownin this tissue (Je et al., 2001). Finally, even though (asmentioned above) total ERK1/2 protein levels do not changewith antisense treatment, SmAV knockdown inhibits PKC-mediated activation of both ERK1 and ERK2 (Fig. 9E), asdetected with an anti-phospho-ERK1/2 antibody. A similarinhibition of PKC-induced ERK1/2 activation has also beenobserved after CaP knockdown (Je et al., 2001). These resultsimplicate SmAV in both PKC-dependent contractility andsignaling to ERK1/2.

Localization to the cell cortexTo define further the intracellular site(s) at or near the plasmamembrane to which SmAV and CaP relocate after activation ofPKC, we immunolocalized endogenous levels of these proteinsin cultured A7r5 vascular smooth muscle cells before (Fig.10A) and after (Fig. 10B) treatment with a phorbol ester,PDBu. The spread morphology of these cells facilitatesresolution of subcellular localizations especially betweendiverse membrane domains. Phorbol-ester treatment of thesecells induces a disassembly of actin filament bundles and theconcomitant appearance of podosome-like, tubular structurescontaining F-actin, myosin II and other cytoskeletal proteinsthat arise from the ventral cell surface and protrude severalmicrometers into the cell cytoplasm (Brandt et al., 2002; Fultzand Wright, 2003; Hai et al., 2002; Kaverina et al., 2003). Inuntreated cells, SmAV is localized to membrane ruffles andnon-filamentous structures in the cell interior (Fig. 10Aa). CaPis localized primarily to F-actin-containing filaments (Fig.10Ag,k). In PDBu-treated A7r5 cells (Fig. 10B), we found thatendogenous SmAV colocalizes with F-actin (Fig. 10Ba-c) andendogenous h1 CaP (Fig. 10Bg-i) at podosomes (arrowheads)and membrane ruffles (arrows). Podosomes also containnonmuscle myosin IIB (Fig. 10Bd-f) and ERK1/2 (Fig. 10Bj-l). Because the antibodies used in this study are specific onimmunoblots (not shown), these results suggest that PKCinduces the relocation of SmAV, CaP, nonmuscle myosin IIBand ERK1/2 to podosome-like structures at or near theplasma membranes of vascular smooth-muscle cells. Thephenylephrine-induced shifts in protein distributions in A7r5cells (Fig. 10) are qualitatively similar to those observed inthe differentiated aortic smooth-muscle cells (Figs 7, 8),suggesting that the peripheral distribution of SmAV and CaPin the aortic cells might arise from an abundance of similarpodosome-like structures (i.e. complexes containing bothmembrane and cytoskeletal components).

DiscussionThis paper describes for the first time a smooth-muscle isoformof supervillin. This isoform (SmAV) was identified as aninteracting protein with CaP in a yeast two-hybrid assay. Wehave verified this interaction using both in vitro and in vivotechniques, including functional assays. SmAV appears to bethe predominant supervillin/archvillin homolog in smoothmuscle, because western blots of smooth-muscle homogenateswith an antibody against supervillin show only a single bandat the expected molecular mass of SmAV. Furthermore, only asingle RT-PCR product was found for each primer pair duringthe cloning of the cDNA.

It is noteworthy that the SmAV sequence contains severalpredicted nuclear targeting sequences and yet imaging ofdifferentiated smooth muscle cells clearly indicates an absenceof SmAV from the nucleus in both stimulated and unstimulatedcells. By contrast, supervillin (Pestonjamasp et al., 1997) andskeletal-muscle archvillin (Oh et al., 2003) can exhibit strongnuclear localization. This apparent difference in the isoformsmight be explained by a retention of SmAV in the cytoplasmby virtue of its interaction with CaP and/or with other smooth-muscle cytoskeletal elements.

SmAV associates with CaP biochemically and functionally.We observe SmAV-CaP interactions in reciprocal yeast two-

Fig. 8. Quantification of subcellular distribution of CaP and SmAV indifferentiated smooth-muscle cells. Fluorescence intensities werequantified from confocal micrographs. In all cases, between 14 and53 cells were analysed. (Inset) Method used to produce line scans forquantification of ratios. (A) Comparison of the average (withstandard error) of surface:core ratios for the distribution of SmAV inthe absence and presence of phenylephrine (PE). (B) Comparison ofthe average surface:core ratios for the distribution of CaP in theabsence and presence of PE. (C) Average ratios (+ s.e.m.) of SmAVsurface-membrane to cytosol fluorescence intensities in singlesmooth-muscle cells that were unstimulated or stimulated with PEfor 2 minutes, 4 minutes or 10 minutes. ****P<0.0001 comparedwith unstimulated values.

5054

hybrid assays, direct binding on reciprocal blot overlays andcolocalization in doubly transfected COS-7 cells. EndogenousSmAV and CaP co-translocate with similar kinetics in responseto PKC-stimulating agonists in freshly isolated cells fromsmooth muscle, and are co-extracted in the detergent-resistantcytoskeletal fraction from this tissue. Endogenous SmAV andCaP also are co-targeted with ERK1/2 to PKC-induced‘podosome-like’ structures in A7r5 cells. Finally, antisenseknockdown of SmAV inhibits stimulus-mediated contractionthrough a signaling pathway that is indistinguishable from thatpreviously described for CaP (Je et al., 2001).

Because CaP and supervillin/archvillin both contain bindingsites for F-actin and/or myosin II (Chen et al., 2003; Gimonaet al., 2002; Gusev, 2001), it is important to realize that theinteraction described here does not require the presence ofeither actin or myosin. Interactions were observed both in vitrowith purified recombinant proteins and in vivo, when CaP wasco-transfected into COS-7 cells with an EGFP-tagged chimericprotein containing the CaP-binding domain of supervillin butlacking interaction sites for F-actin or myosin II.

By contrast, the co-fractionation of SmAV and CaP in acytoskeletal/detergent-resistant fraction in both resting andstimulated tissues is consistent with multiple cytoskeletalassociations and might reflect a regulatory role for theseproteins during actin- and myosin-mediated contractility.Detergent-resistant cell fractions contain many components,including cholesterol-rich membrane domains such asthose containing caveolin and other lipid-raft-organizingcomponents, as well as cytoskeletal proteins. In fact,

supervillin and skeletal-muscle archvillin associate closelywith both lipid-raft-organizing proteins and cytoskeletalelements (Nebl et al., 2002).

The agonist-stimulated translocation of endogenous SmAVand CaP to the cell periphery in both differentiated smooth-muscle cells and A7r5 cells further suggests a role for theseproteins in mediating interactions between the corticalmembrane skeleton and the contractile apparatus. CaP haspreviously been reported to translocate from contractilefilaments containing α-actin to the β-actin containing networkof the submembranous cortex (Parker et al., 1998). Thus, wesuggest that CaP and SmAV translocate from different siteswithin the cell to a common cytoskeletal structure that is incontact with a detergent-resistant component of the cellmembrane.

The peripheral localization of CaP and SmAV indifferentiated smooth-muscle cells might reflect a recruitmentinto structures analogous to the ‘podosome-like structures’ inimmortalized A7r5 cultured cells, to which CaP and SmAValso colocalize after PKC activation by PDBu. Podosomes areactin- and myosin-II-associated tubular invaginations of thebasal surface of the plasma membrane that have beenimplicated in remodeling of the cortical actin cytoskeleton(Brandt et al., 2002; Fultz et al., 2000; Gimona et al., 2003;Fultz and Wright, 2003; Li et al., 2001). They also contain α-actinin, vinculin and transfected h2 CaP (Gimona et al., 2003;Hai et al., 2002). However, in the same study (Gimona et al.,2003), h1 CaP transfected into these cells was not detected inpodosomes even though displacement of CaP from actin

Journal of Cell Science 117 (21)

Fig. 9. Antisense knockdown of SmAVdecreases contractility and signaling in smooth-muscle tissue. (A) Densitometric analysis ofSmAV protein levels from immunoblots ofantisense- and sham-treated muscles expressedas a precentage of random loaded tissue.** P<0.01. (Inset) A typical western blot forSmAV. (B) Magnitude of steady-statecontraction in response to a depolarizingphysiological saline solution containing 51 mMKCl. Forces are normalized to the amplitude ofcontraction of each muscle in response to 51mM KCl physiological saline solution on day 1.(C) Contraction of muscle strips in response tothe phorbol ester DPBA (3 µM). Forces arenormalized to the amplitude of contraction ofeach muscle in response to 51 mM KCl on day1. **P<0.01 for antisense compared with sham;++P<0.001 for antisense compared withrandom. (D) Contraction of muscle strips inresponse to 10 µM phenylephrine in the absenceof extracellular calcium. Forces are normalizedto the amplitude of contraction of each muscle inresponse to 51 mM KCl physiological saline onday 1. *P<0.05 for antisense compared withsham; +P<0.05 for antisense compared withrandom. (E) Densitometric analysis of phospho-ERK1/2 on immunoblots for antisense- or sham-treated muscles exposed to DPBA, normalized tothat for muscles treated with a random sequence.(Inset) A typical western blot for phospho-ERK1/2. **P<0.05. All values are frombetween three and five separate experiments.

5055Smooth muscle archvillin

filaments is thought to be a prerequisite for podosomeformation (Burgstaller and Gimona, 2004). Interestingly, wefound that endogenous h1 CaP is indeed localized to thepodosomes. Perhaps the dynamics of h1 CaP distributionchange when h1 CaP levels increase after transfection.

This is the first report of which we are aware for thelocalization of supervillin, archvillin, endogenous h1 CaP orERK1/2 to these podosome-like structures. Hai et al. havesuggested that podosomes represent molecular scaffolds forPKC-mediated effects leading to Ca2+ sensitization in smoothmuscle (Hai et al., 2002). Because the targeting of ERK to thecell periphery is PKC dependent and associated with Ca2+

sensitization in smooth muscle (Collins et al., 1992; Khalil etal., 1995), our demonstration of PDBu-mediated recruitmentof ERK1/2 to podosomes supports this hypothesis.

The specific interaction between SmAV and the CH domainof CaP is interesting because this domain is known to bindERK (Leinweber et al., 1999). However, the N-terminalend of SmAV is predicted to contain ERK binding andphosphorylation sites. This raises the possibility that thebinding of the CH domain of CaP to SmAV might displace

ERK1/2 from CaP and free ERK to interact with SmAV orother proteins in the podosome.

The question arises of the mechanism by which CaP/SmAVdownregulation leads to the observed inhibition of contractility.CaP has been previously suggested to function as an adapterprotein to facilitate the targeting of PKC and ERK1/2 to asignaling complex in the cell cortex (Leinweber et al., 2000;Menice et al., 1997). Once CaP, ERK1/2 and SmAV arrive atthe cell cortex, SmAV might displace ERK1/2 from CaP,allowing ERK to be activated by the upstream kinase, MEK.Subsequent ERK1/2-mediated phosphorylation of caldesmonat Ser-789 might then relieve the caldesmon-mediatedinhibition of actomyosin interactions (Sobue et al., 1982). Inthis model, the absence of either SmAV or CaP inhibits ERK-dependent effects but has little or no effect on contractilitytriggered by calcium-dependent signaling events. It is worthnoting that the peak of myosin light chain phosphorylation inthis tissue occurs by 1 minute after phenylephrine stimulationand then declines (Menice et al., 1997), preceding the maximaltranslocation of SmAV (at 4 minutes). Thus myosinphosphorylation pathways might predominate during the

Fig. 10. Localization of endogenous SmAV, F-actin, nonmuscle myosin II, CaP and ERK1/2 before and after PDBu treatment of A7r5 smooth-muscle cells. (A) In untreated A7r5 cells, endogenous SmAV (a) appears to be membranous, whereas F-actin (b,e,h), nonmuscle myosin II (d)and CaP (g,k) are primarily associated with microfilaments, and ERK1/2 appears to be cytosolic (j). (B) After PKC activation with PDBu,endogenous SmAV (a,g), F-actin (b,e), nonmuscle myosin II (d), CaP (h,k) and ERK1/2 (j) all localize to podosome-like structures(arrowheads). SmAV, F-actin, myosin IIB and CaP also concentrate at or near membrane ruffles (arrows). Merged images (c,f,i,l) show theoverlap of signal from panels on the left (green) with the signal from the middle panels (red); the overlap appears yellow. Scale bar, 25 µm.

5056

onset of smooth muscle contractions, whereas caldesmonphosphorylation might play a greater role in the sustainedincrease in contractility. The presence of predicted ERK-binding motifs and proline-directed phosphorylation sites inSmAV further suggests that interactions with membrane and/orcytoskeletal proteins might be under feedback regulatorycontrol from an ERK pathway.

Finally, the phorbol-ester-induced colocalization of CaP andSmAV with ERK to podosomes might imply a role for theseor analogous cortical signaling structures in the regulation ofMAPKs in general. Thus, this paper might provide a ‘missinglink’ between previous work documenting the importance ofthe MAPK cascade to early myogenesis of C2C12 cells(Bennett and Tonks, 1997; Cuenda and Cohen, 1999; Westonet al., 2003) and results showing a dominant-negative effect ofarchvillin sequences on C2C12 myogenesis (Oh et al., 2003).Thus, the stimulus-triggered targeting of CaP and SmAV tothe cell cortex might participate in both the regulation ofcontractility and the coordination of cortical signalingpathways in differentiated and proliferative muscle cells.

We thank W. Marganski (Boston Biomedical Research Institute,Watertown, MA) for colocalization analysis. This research wassupported by NIH grants HL31704, HL42293 and HD3054 toKGM, NIH grant HL074470 to SGG and an MDA award and anNIH grant GM33048 to EJL. We thank Z. Grabarek (BostonBiomedical Research Institute) for the gift of recombinant humancalmodulin, expressed in E. coli and M. Ikebe (University ofMassachusetts Medical School, Worcester, MA) for a critical reviewof the manuscript. We also thank J. Crowley and S. Palmieri forhelpful suggestions on the manuscript, and L. Ohrn for solutionpreparation.

ReferencesAltschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller,

W. and Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a newgeneration of protein database search programs. Nucleic Acids Res.25,3389-3402.

Bennett, A. M. and Tonks, N. K. (1997). Regulation of distinct stages ofskeletal muscle differentiation by mitogen-activated protein kinases. Science278, 1288-1291.

Brandt, D., Gimona, M., Hillmann, M., Haller, H. and Mischak, H. (2002).Protein kinase C induces actin reorganization via a Src- and Rho-dependentpathway. J. Biol. Chem.277, 20903-20910.

Burgstaller, G. and Gimona, M. (2004). Actin cytoskeleton remodelling vialocal inhibition of contractility at discrete microdomains. J. Cell Sci.117,223-231.

Carpenter, G. (2000). The EGF receptor: a nexus for trafficking and signaling.BioEssays22, 697-707.

Chen, Y., Takizawa, N., Crowley, J. L., Oh, S. W., Gatto, C. L., Kambara,T., Sato, O., Li, X.-D., Ikebe, M. and Luna, E. J. (2003). F-Actin andmyosin II binding domains in supervillin. J. Biol. Chem.278, 46094-46106.

Collins, E. M., Walsh, M. P. and Morgan, K. G. (1992). Contraction of singlevascular smooth muscle cells by phenylephrine at constant [Ca2+]i. Am. J.Physiol. Heart Circ. Physiol.262, H754-H762.

Cuenda, A. and Cohen, P. (1999). Stress-activated protein kinase-2/p38 anda rapamycin-sensitive pathway are required for C2C12 myogenesis. J. Biol.Chem.274, 4341-4346.

Feliciello, A., Gottesman, M. E. and Avvedimento, E. V. (2001). Thebiological functions of A-kinase anchor proteins. J. Mol. Biol.308, 99-114.

Fultz, M. E. and Wright, G. L. (2003). Myosin remodelling in the contractingA7r5 smooth muscle cell. Acta Physiol. Scand.177, 197-205.

Fultz, M. E., Li, C., Geng, W. and Wright, G. L. (2000). Remodeling of theactin cytoskeleton in the contracting A7r5 smooth muscle cell. J. MuscleRes. Cell Motil.21, 775-787.

Ge, L., Ly, Y., Hollenberg, M. and DeFea, K. (2003). A beta-arrestin-dependent scaffold is associated with prolonged MAPK activation in

pseudopodia during protease-activated receptor-2-induced chemotaxis. J.Biol. Chem.278, 34418-34426.

Gerthoffer, W. T. and Gunst, S. J. (2001). Focal adhesion and small heatshock proteins in the regulation of actin remodeling and contractility insmooth muscle. J. Appl. Physiol.91, 963-972.

Gimona, M., Djinovic-Carugo, K., Kranewitter, W. J. and Winder, S. J.(2002). Functional plasticity of CH domains. FEBS Lett.513, 98-106.

Gimona, M., Kaverina, I., Resch, G. P., Vignal, E. and Burgstaller, G.(2003). Calponin repeats regulate actin filament stability and formation ofpodosomes in smooth muscle cells. Mol. Biol. Cell14, 2482-2491.

Gusev, N. B. (2001). Some properties of caldesmon and calponin and theparticipation of these proteins in regulation of smooth muscle contractionand cytoskeleton formation. Biochemistry66, 1112-1121.

Hai, C. M., Hahne, P., Harrington, E. O. and Gimona, M. (2002).Conventional protein kinase C mediates phorbol-dibutyrate-inducedcytoskeletal remodeling in A7r5 smooth muscle cells. Exp. Cell Res.280,64-74.

Harnett, K. M. and Biancani, P. (2003). Calcium-dependent and calcium-independent contractions in smooth muscles. Am. J. Med.115, S24-S30.

Je, H. D., Gangopadhyay, S. S., Ashworth, T. D. and Morgan, K. G. (2001).Calponin is required for agonist-induced signal transduction – evidencefrom an antisense approach in ferret smooth muscle. J. Physiol.537, 567-577.

Jiang, M. J. and Morgan, K. G. (1989). Agonist-specific myosinphosphorylation and intracellular calcium during isometric contractions ofarterial smooth muscle. Pflügers Archiv.413, 637-643.

Kaverina, I., Stradal, T. E. and Gimona, M. (2003). Podosome formationin cultured A7r5 vascular smooth muscle cells requires Arp2/3-dependentde-novo actin polymerization at discrete microdomains. J. Cell Sci.116,4915-4924.

Khalil, R. A., Lajoie, C. and Morgan, K. G. (1994). In situ determination ofthe [Ca2+]i threshold for translocation of the alpha protein kinase C isoform.Am. J. Physiol., Cell Physiol.266, C1544-C1551.

Khalil, R. A., Menice, C. B., Wang, C.-L. A. and Morgan, K. G. (1995).Phosphotyrosine-dependent targeting of mitogen-activated protein kinase indifferentiated contractile vascular cells. Circ. Res.76, 1101-1108.

Kim, I., Je, H. D., Gallant, C., Zhan, Q., Riper, D. V., Badwey, J. A., Singer,H. A. and Morgan, K. G. (2000). Ca2+-calmodulin-dependent proteinkinase II-dependent activation of contractility in ferret aorta. J. Physiol.526,367-374.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assemblyof the head of bacteriophage T4. Nature227, 680-685.

Leinweber, B. D., Leavis, P. C., Grabarek, Z., Wang, C.-L. A. and Morgan,K. G. (1999). Extracellular regulated kinase (ERK) interaction with actinand the calponin homology (CH) domain of actin-binding proteins.Biochem. J.344, 117-123.

Leinweber, B., Parissenti, A. M., Gallant, C., Gangopadhyay, S. S.,Kirwan-Rhude, A., Leavis, P. C. and Morgan, K. G. (2000). Regulationof protein kinase C by the cytoskeletal protein calponin. J. Biol. Chem.275,40329-40336.

Li, C., Fultz, M. E., Parkash, J., Rhoten, W. B. and Wright, G. L. (2001).Ca2+-dependent actin remodeling in the contracting A7r5 cell. J. MuscleRes. Cell Motil.22, 521-534.

Menice, C. B., Hulvershorn, J., Adam, L. P., Wang, C. A. and Morgan,K. G. (1997). Calponin and mitogen-activated protein kinase signalingin differentiated vascular smooth muscle. J. Biol. Chem.272, 25157-25161.

Miller, W. E. and Lefkowitz, R. J. (2001). Expanding roles for beta-arrestinsas scaffolds and adapters in GPCR signaling and trafficking. Curr. Opin.Cell Biol. 13, 139-145.

Moreland, S. J., Nishimura, J., van Breemen, C., Ahn, H. Y. andMoreland, R. S. (1992). Transient myosin phosphorylation at constant Ca2+

during agonist activation of permeabilized arteries. Am. J. Physiol.263,C540-C544.

Morgan, K. G. and Gangopadhyay, S. S. (2001). Cross-bridge regulation bythin filament-associated proteins. J. Appl. Physiol.91, 953-962.

Nebl, T., Pestonjamasp, K. N., Leszyk, J. D., Crowley, J. L., Oh, S. W. andLuna, E. J. (2002). Proteomic analysis of a detergent-resistant membraneskeleton from neutrophil plasma membranes. J. Biol. Chem.277, 43399-43409.

Oh, S. W., Pope, R. K., Smith, K. P., Crowley, J. L., Nebl, T., Lawrence,J. B. and Luna, E. J. (2003). Archvillin, a muscle-specific isoform ofsupervillin, is an early expressed component of the costameric membraneskeleton. J. Cell Sci.116, 2261-2275.

Journal of Cell Science 117 (21)

5057Smooth muscle archvillin

Parker, C. A., Takahashi, K., Tang, J. X., Tao, T. and Morgan, K. G.(1998). Cytoskeletal targeting of calponin in differentiated, contractilesmooth muscle cells of the ferret. J. Physiol.508, 187-198.

Pestonjamasp, K., Amieva, M. R., Strassel, C. P., Nauseef, W. M.,Furthmayr, H. and Luna, E. J. (1995). Moesin, ezrin, and p205 are actin-binding proteins associated with neutrophil plasma membranes. Mol. Biol.Cell 6, 247-259.

Pestonjamasp, K. N., Pope, R. K., Wulfkuhle, J. D. and Luna, E. J. (1997).Supervillin (p205): a novel membrane-associated, F-actin-binding protein inthe villin/gelsolin superfamily. J. Cell Biol.139, 1255-1269.

Pfitzer, G. (2001). Regulation of myosin phosphorylation in smooth muscle.J. Appl. Physiol.91, 497-503.

Pouysségur, J. and Lenormand, P. (2003). Fidelity and spatio-temporalcontrol in MAP kinase (ERKs) signalling. Eur. J. Biochem.270, 3291-3299.

Pryciak, P. M. (2001). MAP kinases bite back. Dev. Cell1, 449-451.Sim, A. T. and Scott, J. D. (1999). Targeting of PKA, PKC and protein

phosphatases to cellular microdomains. Cell Calcium26, 209-217.Small, J. V. and Gimona, M. (1998). The cytoskeleton of the vertebrate

smooth muscle cell. Acta Physiol. Scand.164, 341-348.Sobue, K., Morimoto, K., Inui, M., Kanda, K. and Kakiuchi, S. (1982).

Control of actin-myosin interaction of gizzard smooth muscle by calmodulinand caldesmon-linked flip-flop mechanism. Biomed. Res.3, 188-196.

Somlyo, A. P. and Somlyo, A. V. (2003). Ca2+ sensitivity of smooth muscleand nonmuscle myosin II: modulated by G proteins, kinases, and myosinphosphatase. Physiol. Rev.83, 1325-1358.

Swaffield, J. C. and Johnston, S. A. (2003). Affinity Purification of ProteinsBinding to GST Fusion Proteins. Indianapolis, IN: John Wiley & Sons.

Takahashi, K., Hiwada, K. and Kokubu, T. (1986). Isolation andcharacterization of a 34000-Dalton calmodulin- and F-actin-binding proteinfrom chicken gizzard smooth muscle. Biochem. Biophys. Res. Commun.141, 20-26.

Ting, H. J., Yeh, S., Nishimura, K. and Chang, C. (2002). Supervillinassociates with androgen receptor and modulates its transcriptional activity.Proc. Natl. Acad. Sci. USA 99, 661-666.

Wang, C. L. (2001). Caldesmon and smooth-muscle regulation. Cell Biochem.Biophys.35, 275-288.

Weston, A. D., Sampaio, A. V., Ridgeway, A. G. and Underhill, T. M.(2003). Inhibition of p38 MAPK signaling promotes late stages ofmyogenesis. J. Cell Sci.116, 2885-2893.

Wier, W. G. and Morgan, K. G. (2003). alpha(1)-Adrenergic signalingmechanisms in contraction of resistance arteries. Rev. Physiol. Biochem.Pharmacol. 150, 91-139.

Winder, S. J., Allen, B. G., Clement-Chomienne, O. and Walsh, M. P.(1998). Regulation of smooth muscle actin-myosin interaction and force bycalponin. Acta Physiol. Scand.164, 415-426.

Wulfkuhle, J. D., Donina, I. E., Stark, N. H., Pope, R. K., Pestonjamasp,K. N., Niswonger, M. L. and Luna, E. J. (1999). Domain analysis ofsupervillin, an F-actin bundling plasma membrane protein with functionalnuclear localization signals. J. Cell Sci.112, 2125-2136.

Zamir, E. and Geiger, B. (2001). Molecular complexity and dynamics of cell-matrix adhesions. J. Cell Sci.114, 3583-3590.